How an instrument for measuring the health of civilisations got built — and what it has found, so far, by pointing itself at real history.
A commentator says a country is "fracturing," or "resilient," or "in decline," and there is usually no way to test the claim against a rival's claim, or against the same country twenty years earlier, because nobody wrote down what they were actually measuring. CAMS — the Complex Adaptive Model of Societies — started from the suspicion that this doesn't have to be true. If a society is a coordination system, its coordination should be measurable the same way in every era and every country, which means claims about it should be falsifiable the same way a claim in any other science is.
Recorded history, on this view, is sediment: every crisis, reform, and rupture a society has lived through is still legible in how its institutions behave later, the way weather is legible in tree rings. CAMS tries to core-sample that record directly — not narrate it, measure it.
Any group of humans coordinating at scale — a village, a corporation, a nation — appears to need the same eight things done, whatever it calls them. CAMS scores each of the eight on four dimensions every year: how internally aligned it is (Coherence), how much real capability it commands (Capacity), how much pressure it's under (Stress), and how far its official story has drifted from its material reality (Abstraction).
A score is not read off a spreadsheet the way GDP is. It is a judgment — rendered by independent scorers reading the historical record for that year — which is why CAMS runs five scoring passes per period rather than one, and reports the disagreement between them as part of the result rather than hiding it. Where the evidence is too thin to support a confident number, the newer scoring protocols record that honestly as insufficient evidence rather than guessing.
An old question, asked of a new collaborator
How do you tell knowledge from persuasion — how do you stop a political story from grading its own truth? The project began as that question, put to an AI system, rather than as an equation.
First formal definition
Published under the name Common Adaptive Model State. Eight roles, described informally — Leadership, Warrior Class, Priesthood, and so on — scored on three dimensions, the third then called "Impact."
CAMS, and a fourth dimension
"Impact" was replaced by Abstraction once it became clear that what separated resilient societies from brittle ones was internal symbolic honesty, not external reach. The short name "CAMS" enters public use alongside the first scored comparison: Russia, China, the USA, and Singapore, 2000–2025.
Functional renaming
The eight roles are renamed to Helm, Shield, Lore, Archive, Craft, Flow, Stewards, and Hands — functions any durable collective must solve, rather than roles tied to one historical class structure, so the same instrument can be pointed at any scale.
JUNO and system-level diagnostics
A further layer adds whole-system health measures and named regimes a society can occupy in a given year — Stable Adaptive, Strained, Fracture — plus a formal, blind, 229-polity clustering pass to test whether civilisational "types" hold up statistically rather than by assertion.
An instrument earns trust on a disputed question partly by first finding, cleanly and without being told where to look, the crises nobody disputes.
The Rum Rebellion registers as the single sharply negative Node Value anywhere in Australia's 200-year record, recovering fully within two years under Macquarie — a governance collapse and rebound the historical record already agrees on.
A sealed calibration run against the Civil War passed all eight of its pre-registered structural expectations, correctly isolating the Helm–Shield coordination failure historians already identify as the war's institutional core.
Both nations' occupation years surface, independently, as the deepest trough in their entire measured history — Singapore's alone severe enough to register the model's rarest classification, full systemic collapse — each followed by a full and rapid recovery.
Run across enough countries, one pattern keeps recurring: a society's material and institutional capacity can stay strong for decades after its capacity for coherent governance and shared meaning has begun a slow decline — the two are not the same thing, and CAMS was built specifically to stop conflating them.
The most striking single result, though, sits above any one country. A clustering pass across 229 historical and modern polities — blind to geography, ideology, and regime type — grouped modern China and modern Singapore with Rome at its height, the Ottoman Empire, and the British Empire, on shared structural grounds the source material calls explicitly "a structural finding and not an ideological one." A method built to test intuitions rather than confirm them is only doing its job when it occasionally says something nobody was trying to hear.
Something moves the troop beneath the stories the troop tells about itself.